github/tinygrad
1
1
Fork 0
mirror of https://github.com/tinygrad/tinygrad.git synced 2026-01-26 07:18:40 -05:00
Code Issues Packages Projects Releases Wiki Activity
2,388 Commits 206 Branches 14 Tags
f9cb31fdc2c8df2fdb84873924b6dbd8e67cd315
Commit Graph

2388 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Max Hahn
f9cb31fdc2 added visitor pattern (#1669) 
* added visitor pattern

* pylint bug workaround

* added tests, made abstract OpNode inherit from ABC

* fixed assert

* fix check of abstract classes in negative test

* remove assert False
 2023-08-30 09:03:44 -07:00
George Hotz
fdd7f282cb Reenable tensor cores for self-hosted Mac CI (#1717) 
* debug 5 matmul

* allow tensor cores in CI

* tensor cores on arm64

* put debug back
 2023-08-30 07:53:04 -07:00
chenyu
ac183568be llama JIT python runtime speedup (#1633) 
* no JIT call in TransformerBlock

* idea

* move 2 reshapes to jitted function

shrink inside jitted too, 6.3ms

remove back reshapes, 5.5ms

isinstance -> __class__ 4.99ms

* think

revert ops_gpu.py

revert symbolic.py too

PYOPENCL_COMPILER_OUTPUT=1

* cleanup

* fix cache shape for conversational model

only reshape if start_pos > 0

* small cleanup

* include var_vals.keys() to st.key

* add comments

* llama small update

* everything jitted again, similar structure to gpt2

* fix typing

* add TODO for in place update cache
 2023-08-30 07:51:05 -07:00
Umut Zengin
1682e9a38a Fix: Stable Diffusion index (#1713) 2023-08-30 00:21:10 -04:00
wozeparrot
2f768e386d stable diffusion benchmark artifact (#1714) 2023-08-29 21:08:40 -04:00
George Hotz
0ea22bf249 remove DEBUG=1 from stable diffusion AMD since jit cache is fixed 2023-08-29 12:46:12 -07:00
George Hotz
ab9b9ff3e2 pipefail benchmark (#1709) (#1710) 
* feat: specify shell

* feat: specify shell for mac

Co-authored-by: wozeparrot <wozeparrot@gmail.com>
 2023-08-29 08:15:02 -07:00
George Hotz
aa7c98722b sd timing (#1706) 2023-08-28 20:22:57 -07:00
nimlgen
8844a0a822 llvm jitted (#1652) 2023-08-28 20:22:44 -07:00
nimlgen
1c0449e190 add cache collector (#1595) 
* init cache collector

* add test_cache_collector.py

* switch GlobalCounters.cache to CacheCollector

* init jit models test

* jitted SD

* add debug msg to print loaded bufs count

* moved cache collctor to jit

* clearer SD

* no double device import
 2023-08-28 19:59:55 -07:00
George Hotz
f5f8b09c13 allow manual release (#1704) 2023-08-28 17:54:25 -07:00
George Hotz
715047a1e4 fix release publish (#1703) 2023-08-28 17:48:00 -07:00
Olivier Chafik
ee6d8de2dc Llama: load models in HuggingFace format (incl. indexed, safetensors) (#1583) 2023-08-28 15:11:40 -04:00
qazal
3515ba4f23 add dtypes test (#1682) 2023-08-28 08:12:15 -07:00
Roelof van Dijk
50f669e43b [ready] perf: simpler Tensor init (#1679) 
Co-authored-by: Roelof van Dijk <roelof.van.dijk@vitestro.com>
 2023-08-27 22:18:03 -04:00
Roelof van Dijk
b66f54e379 perf: avoid reshaping if not necessary (#1683) 
Co-authored-by: Roelof van Dijk <roelof.van.dijk@vitestro.com>
 2023-08-27 14:17:04 -04:00
Roelof van Dijk
328cf2e86a perf: remove cast and revert back to isinstance (#1694) 
Co-authored-by: Roelof van Dijk <roelof.van.dijk@vitestro.com>
 2023-08-27 14:15:52 -04:00
wozeparrot
8b354b3f73 feat: version bump! (#1687) v0.7.0 2023-08-27 12:38:58 -04:00
Roelof van Dijk
abaa605f71 [ready] perf: start enumerate at 1 instead of checking all i (#1691) 
Co-authored-by: Roelof van Dijk <roelof.van.dijk@vitestro.com>
 2023-08-27 12:00:32 -04:00
Roelof van Dijk
2730ed657f perf: faster lazyop eq (#1693) 
Co-authored-by: Roelof van Dijk <roelof.van.dijk@vitestro.com>
 2023-08-27 11:17:02 -04:00
Roelof van Dijk
6ca509a485 perf: constant in while in for in busy func (#1688) 
Co-authored-by: Roelof van Dijk <roelof.van.dijk@vitestro.com>
 2023-08-27 11:13:16 -04:00
Roelof van Dijk
b89d81330f fix: restore old behaviour (#1689) 
Co-authored-by: Roelof van Dijk <roelof.van.dijk@vitestro.com>
 2023-08-27 10:45:53 -04:00
chenyu
66fbf4800b fix symbolic_ops tests with Tensor.training=True (#1686) 2023-08-26 23:19:56 -04:00
Roelof van Dijk
6c5dc9c153 [ready] perf: faster lazyop init (#1673) 
Co-authored-by: Roelof van Dijk <roelof.van.dijk@vitestro.com>
 2023-08-26 22:59:10 -04:00
wozeparrot
f61d0657d1 document new envvars (#1676) 
* feat: document some new envvars

* feat: actually put values

* feat: no more cifar torch

* feat: no fakedata
 2023-08-26 20:17:02 -04:00
Yixiang Gao
9d93a82354 remove FAKEDATA (#1685) 2023-08-26 20:15:54 -04:00
chenyu
b5d700adae update openpilot supercombo.onnx to 0.9.4 (#1681) 
* update openpilot supercombo.onnx to 0.9.4

* update tests for the new model

* comment out comma models from external_model_benchmark
 2023-08-26 19:16:08 -04:00
Roelof van Dijk
89b529c07f [ready] ci: add py38 to linters (#1674) 
* ci: add py38 to linters

* fix: run linters only on py38

---------

Co-authored-by: Roelof van Dijk <roelof.van.dijk@vitestro.com>
 2023-08-26 09:34:15 -04:00
Jordan Wright
25be7f745d Tensor.uniform with dtype=int bug fix (#1593) 2023-08-26 01:59:53 -04:00
Roelof van Dijk
f702a8f497 [ready] avoid in-function graph imports in lazy.py (#1666) 
Co-authored-by: Roelof van Dijk <roelof.van.dijk@vitestro.com>
 2023-08-25 13:56:28 -04:00
Roelof van Dijk
02e64da678 refactor: tuples can be concatenated with + (#1671) 
Co-authored-by: Roelof van Dijk <roelof.van.dijk@vitestro.com>
 2023-08-25 12:37:13 -04:00
Yixiang Gao
173850f599 fix CIFAR jit (#1657) 
* update mask function

* kept 94 with the new fetcher

clean up batch fetcher

* 94.04% without cutmix

* 94.04% with cutmix

* move batch fetcher to avoid fetching additional batch last STEP
 2023-08-24 16:14:40 -07:00
chenyu
f00325e77d ops_metal newCommandQueueWithMaxCommandBufferCount_(1024) (#1664) 2023-08-24 15:42:00 -07:00
DavidFarago
1ba8f0dca3 Quickstart: Upgrade section "Training" to new code (#1663) 
Co-authored-by: Dave Farago <dfarago@innoopract.com>
 2023-08-24 17:12:16 -04:00
DavidFarago
29adae84eb Quickstart: Use tensors to compute train accuracy (#1662) 
Co-authored-by: Dave Farago <dfarago@innoopract.com>
 2023-08-24 17:09:12 -04:00
George Hotz
d37d092c14 split linearizer into 3 files (#1654) 2023-08-23 14:58:47 -07:00
George Hotz
1b8c40234f Uast start (#1650) 
* work

* more tests

* more tests 2

* don't break it
 2023-08-23 12:00:06 -07:00
geohotstan
484708da87 #1615 fix (#1616) 2023-08-23 14:51:05 -04:00
Pavol Rusnak
b57c374164 add accelerator links to readme (#1649) 2023-08-23 14:47:55 -04:00
George Hotz
82623697a8 Move asm renderer (#1648) 
* teeny changes

* teeny updates

* move to renderer
 2023-08-23 10:06:43 -07:00
George Hotz
a89363574d teeny changes (#1647) 
* teeny changes

* teeny updates
 2023-08-23 09:53:39 -07:00
George Hotz
a6d842af7a move device to ops (#1646) 
* move device to ops

* mlops types

* 2 lines
 2023-08-23 08:30:17 -07:00
nimlgen
a65ae1198b do replace div->mul for non-floats (#1644) 2023-08-23 07:34:31 -07:00
George Hotz
da694d4241 move that image import 2023-08-22 21:30:55 -07:00
George Hotz
41e83be3dd simple where broadcast (#1643) 2023-08-22 21:24:49 -07:00
George Hotz
c831218139 Optional: Reduce line count and simplify the LazyBuffer interface (#1642) 
* less lines in lazybuffer, def e

* custom function

* cast

* reorder functions

* lb type
 2023-08-22 21:01:10 -07:00
George Hotz
d25046e66a matvec tests (#1634) 
* matvec tests

* f16

* f16 is broken
 2023-08-22 17:33:58 -07:00
George Hotz
643cbdfd50 make embedding and GPT-2 fast (#1631) 
* make embedding fast

* jit more, variable shape support

* print mem bw
 2023-08-22 15:14:38 -07:00
Niklas D
a7752ad65d Fix link to state.py in quickstart (#1632) 2023-08-22 17:39:30 -04:00
c143
c9c40bb16f Import whole math module in tensor.py (#1628) 2023-08-22 17:07:46 -04:00